Thicker Myelin Tracked Faster Cognitive Decline in Old Age

Jeff's Daily Pick. In donated brains from the Lothian Birth Cohort 1936, faster cognitive decline after 70 went with thicker myelin on large nerve fibres and oligodendrocytes low in the protein NRF2 (Nature Medicine). In mice, removing NRF2 from those cells reproduced the thicker myelin.

In the United Kingdom, researchers at the University of Edinburgh and the UK Dementia Research Institute published a study in Nature Medicine on August 25, 2026 that looks at the brain cells that make myelin, the fatty sheath around nerve fibres. Myelin is usually treated as protective. The paper, led by Georgina Craig with Véronique Miron as senior author, links an excess of it to faster decline in old age. It drew new coverage in October.

A cohort tested since childhood

The work uses the Lothian Birth Cohort 1936: people born in 1936 who sat an intelligence test at about age 11 in 1947, then repeated memory, speed and spatial tests from age 70 to 82, every three years. Of the 866 members retested after 70, 865 declined to some degree. The team sorted them into milder and steeper than average decline, then examined donated brain tissue from the corpus callosum, the main white matter bridge between the two hemispheres.

What the brains showed

Steeper decline went with smaller nerve fibres, more fibres in early degeneration and, unexpectedly, thicker myelin around the large fibres, those over 1 micrometre wide. Compact myelin measured by another method was lower: the paper describes a loss of myelin integrity, not simply more myelin. The changes tracked the rate of decline, not a person's score at any one point. Gene readouts from 45,437 cell nuclei in 14 brains, then staining in 18 to 19, showed more oligodendrocytes, the myelin making cells, and a smaller share of them carrying NRF2, a protein that switches on hundreds of protective genes.

The mouse test

The team then deleted NRF2 only in oligodendrocytes of adult mice, starting at six months of age. In a water maze at 12 months, in 38 male mice, controls improved their time in the target zone by 57.75% over the four test days; the mice without NRF2 improved by 27.82%. Female knockouts showed altered anxiety and were left out of that analysis. At 15 to 18 months, the knockouts had thicker myelin and smaller fibres, the same pattern as in the human brains.

The finding cuts against a common reading of white matter, where loss of myelin is the usual suspect. Here the problem is the opposite on the largest fibres. No trial in people is reported. Our piece on two enzymes and memory in aging mice covers another mouse pathway behind age related memory loss.